Schizophrenia Research xxx (2015) xxx–xxx

SCHRES-06205; No of Pages 6

Contents lists available at ScienceDirect

Schizophrenia Research

j ourna l homepage: www.e lsev ie r .com/ locate /schres

Predictors of neuropsychological effort test performancein schizophrenia

Lindsay F. Morra a, James M. Gold b, Sara K. Sullivan a, Gregory P. Strauss a,⁎a State University of New York at Binghamton, Department of Psychology, USAb University of Maryland School of Medicine, Department of Psychiatry, Maryland Psychiatric Research Center, USA

⁎ Corresponding author at: State University of New YorkPsychology, PO Box 6000, Binghamton, New York 13905408; fax: +1 607 777 4890.

E-mail address: gstrauss@binghamton.edu (G.P. Strau

http://dx.doi.org/10.1016/j.schres.2014.12.0330920-9964/© 2014 Published by Elsevier B.V.

Please cite this article as: Morra, L.F., et al., Phttp://dx.doi.org/10.1016/j.schres.2014.12.03

a b s t r a c t

a r t i c l e i n f o

Article history:Received 6 November 2014Received in revised form 19 December 2014Accepted 22 December 2014Available online xxxx

Keywords:PsychosisEffortCognitionNegative symptomsIntelligence

There is some evidence that insufficient effort may be common in schizophrenia, posing significant threats to thevalidity of neuropsychological test results. Low effort may account for a significant proportion of variance inneuropsychological test scores and the generalized cognitive deficit that characterizes the disorder. The currentstudy evaluated clinical predictors of insufficient effort in schizophrenia using an embedded effort measure, theRepeatable Battery for the Assessment of Neuropsychological Status (RBANS) Effort Index (EI). Participants were330 patients meeting DSM-IV-TR criteria for schizophrenia, schizoaffective disorder, or another psychotic disor-der who received a battery of neuropsychological tests, including: Wechsler Test of Adult Reading (WTAR),Wechsler Abbreviated Scale of Intelligence (WASI), and RBANS. Clinical assessments designed to measure func-tional outcome and symptoms were also obtained. Results indicated that 9.4% of patients failed the EI. Patientswho failed had lower full-scale, verbal, and performance IQ, as well as poorer performance on RBANS domainsnot included in the EI (immediate memory, language, and visuospatial/construction). Patients who failed theEI also displayed poorer community-based vocational outcome, greater likelihood of having “deficit schizophre-nia” (i.e., primary and enduring negative symptoms), and increased severity of positive symptoms. Regressionanalyses revealed that insufficient effort was most significantly predicted by a combination of low IQ, negativesymptoms, and positive symptoms. Findings suggest that although insufficient effort may be relatively uncom-mon in schizophrenia, it is associatedwith important clinical outcomes. The RBANSEImay be a useful tool in eval-uating insufficient effort in schizophrenia.

© 2014 Published by Elsevier B.V.

1. Introduction

Neuropsychological impairment is common in schizophrenia, and haslong been considered a core feature of the illness (Kraepelin, 1919).Meta-analyses indicate that individuals with schizophrenia displayneurocognitive impairments approximately one standard deviationbelow the mean for healthy controls (Fioravanti et al., 2005; Dickinsonet al., 2007). Despite such pervasive cognitive impairments, there is nodistinct pattern of differential deficits that characterizes most individualswith schizophrenia (Reichenberg andHarvey, 2007). Rather, schizophre-nia patients display neurocognitive impairments of similar magnitudeacross most cognitive domains, suggesting a generalized neurocognitivedeficit (Dickinson et al., 2004; Dickinson, 2008; Dickinson et al., 2008).

Several theories have been proposed to account for this generalizedneurocognitive deficit, including central nervous system (e.g., gray andwhite-matter abnormalities, impaired integration of signals across neu-ral networks, and cellular-level neuropathology) and “general systems”

at Binghamton, Department of2-6000, USA. Tel.: +1 607 777

ss).

redictors of neuropsycholog3

(e.g., inflammatory, metabolic, and oxidative stress processes) abnor-malities that can negatively impact cognition (Dickinson and Harvey,2009). However, it is also possible that psychological factors contributesubstantially to the neurocognitive impairments observed in schizo-phrenia. One possibility is that problems with motivation result ininadequate effort on measures of neurocognition, particularly on tasksthat are more cognitively demanding. To date, relatively few studieshave examined insufficient effort during neuropsychological testing inschizophrenia and whether such abnormalities are associated withmotivational problems. Those studies that have been conducted haveproduced inconsistent results, with the majority indicating that asmall proportion of individuals with schizophrenia (~20%) performbelow clinically established cut-off scores for valid effort (Back et al.,1996; Egeland et al., 2003; Arnold et al., 2005; Duncan, 2005; Gieroket al., 2005; Avery et al., 2009; Pivovarova et al., 2009; Schroeder andMarshall, 2011; Moore et al., 2013; Hunt et al., 2014), and other studiesindicating that up to 60–72% of the sample may fail effort testing(Gorissen et al., 2005; Hunt et al., 2014).

Despite these inconsistencies regarding rates of effort test failure,there is reliable evidence that certain clinical variables predict low effortin schizophrenia. For example, multiple studies have found that globalscores on negative symptom rating scales, such as the Scale for the

ical effort test performance in schizophrenia, Schizophr. Res. (2015),

Table 1Participant demographics by group.

SZ-PASS(N = 299)

SZ-FAIL(N = 31)

Test statistic,p value

Age 35.04 (10.36) 41.55 (14.45) F = 10.20, p b 0.01% Male 72.6% 77.4% χ2 = 0.34, p = 0.56Ethnicity χ2 = 1.48, p = 0.83Caucasian 53.8% 61.3%African American 41.8% 32.3%American Indian 0.3% 0%Asian 1.7% 3.2%Other 2.3% 3.2%

Parental education 12.48 (3.11) 11.46 (2.54) F = 2.32, p = 0.13Participant education 12.24 (2.10) 11.38 (2.34) F = 4.39, p = 0.04

Notes: SZ-PASS = participants with an EI N 3, SZ-FAIL, participants with an EI ≤ 3.

2 L.F. Morra et al. / Schizophrenia Research xxx (2015) xxx–xxx

Assessment of Negative Symptoms (Andreasen, 1983) or the BriefNegative Symptom Scale (Kirkpatrick et al., 2011), account for a sub-stantial proportion of variance in effort test performance (Gorissenet al., 2005; Avery et al., 2009; Strauss et al., 2014). Several psychologicalvariables also differentiate patients who pass and fail effort measures,including self-reported anhedonia and the perception of low cognitiveresources (Avery et al., 2009; Strauss et al., 2014). These findings sug-gest that negative symptoms and psychological processes associatedwith negative symptoms may be core to diminished effort during neu-ropsychological testing.

However, negative symptoms are both multi-dimensional andmulti-determined and it is currently unclear which aspects of negativesymptoms are associated with low effort. There is consistent evidencefor themulti-dimensionality of negative symptoms, such that 2 distinctnegative symptom factors are commonly identified, one reflectingdiminished motivation (anhedonia, avolition, asociality) and the otherdiminished expressivity (alogia and restricted affect) (Blanchard andCohen, 2006; Horan et al., 2011; Strauss et al., 2012). These two dimen-sions have different demographic and clinical correlates (Strauss et al.,2013), with more severe volitional pathology generally predictingworse outcomes. Given that neuropsychological impairment has beenassociated with the motivational dimension more strongly than thediminished expressivity dimension (Fervaha et al., 2014), one mightexpect effort test performance to be specifically linked to greater sever-ity ofmotivational symptoms. Furthermore, it is nowgenerally acceptedthat negative symptoms are multi-determined — two patients can dis-play identical scores on negative symptom rating scales for very differ-ent reasons. This notion was highlighted in the seminal work ofCarpenter et al. (1988), which demonstrated that negative symptomscan result from either primary or secondary factors. Primary negativesymptoms are those that are idiopathic to the illness, whereas second-ary negative symptoms result from processes such as paranoid socialwithdrawal, depression, disorganization, hallucinations, and suspi-ciousness. If low effort is indeed critically linked to true motivationalproblems in schizophrenia, one might expect higher rates of effort testfailure in patients who meet clinical diagnostic criteria for “deficitschizophrenia”, i.e., those with primary and clinically stable negativesymptoms (Carpenter et al., 1988; Kirkpatrick et al., 2001). Patientswith deficit schizophrenia typically fall 1 SD below nondeficit schizo-phrenia patients and 2 SD below healthy controls on standard neuro-psychological tests (Buchanan et al., 1994; Cohen et al., 2007);however, it remains to be seen whether patients meeting clinicalcriteria for deficit schizophrenia are more likely to fail effort tests thannondeficit patients.

In the current study, we explored rates of effort test failure in a largesample of individuals with schizophrenia using an embedded effortmeasure that has been well-validated in clinical populations, the RBANSEffort Index (Silverberg et al., 2007). Clinical predictors of insufficienteffort were examined, with an emphasis on determining whether themotivational dimension and primary negative symptoms are most pre-dictive of effort test failure. It was hypothesized that a small percentageof individuals with schizophrenia (b20%) would fail the RBANS EffortIndex and that patients falling below the low-effort cut-off would bemore likely tomeet clinical criteria for deficit schizophrenia, have great-er severity of motivational symptoms, and poorer community-basedfunctional outcome.

2. Method

2.1. Participants

Participants included 330 individuals meeting DSM-III or DSM-IVcriteria for schizophrenia (n = 289), schizoaffective disorder (n = 32),or another psychotic disorder (n = 9). Outpatients were recruitedfrom the Maryland Psychiatric Research Center (MPRC) outpatientclinics and other local outpatient clinical care centers. Inpatients were

Please cite this article as: Morra, L.F., et al., Predictors of neuropsychologhttp://dx.doi.org/10.1016/j.schres.2014.12.033

recruited from the Treatment Research Program unit of the MPRC.Participants were excluded for: (1) history of substance abuse or depen-dence in the past 6 months, (2) history of a head injury, and (3) historyof a neurological disorder. All patientswere assessed in a research (ratherthan clinical) context, and therefore did not have any identifiablemotiva-tion to feign or exaggerate clinical and/or cognitive symptoms. Patientswere not using results of neuropsychological evaluations for disabilitycompensation or litigation purposes.

Participants were divided into those who passed (SZ-PASS) andfailed (SZ-FAIL) the RBANS Effort Index according to established proce-dures (Silverberg et al., 2007). Demographic characteristics of theSZ-PASS and SZ-FAIL groups are presented in Table 1. The two groupsdid not differ on sex, race, or years of parental education. However,the SZ-FAIL groupwas significantly older andhad fewer years of person-al education than SZ-PASS.

2.2. Measures

Participants completed a clinical interview, after which psychiatricrating instruments designed to measure community-based functionaloutcome, psychosis, disorganization, and negative symptoms werecompleted. Neuropsychological tests were also administered. Consen-sus diagnosis was established via a best-estimate approach basedupon multiple interviews and a detailed psychiatric history. Thisdiagnosis was subsequently confirmed using the Structured ClinicalInterview for DSM (SCID). Symptom rating scales were completed byclinicians trained to MPRC reliability standards (reliability N 0.80;Schedule for the Deficit Syndrome kappa N 8/10).

Symptom severity measures included the Schedule for the DeficitSyndrome (SDS: Kirkpatrick et al., 1989) and the Scale for the Assess-ment of Positive Symptoms (SAPS: Andreasen, 1984). The SDS assessesthe severity of six negative symptoms (restricted affect, poverty ofspeech, diminished emotional range, curbed interests, diminishedsense of purpose, and diminished social drive), and whether thosesymptoms are primary/secondary and enduring (stable N 1 year). Themeasure is primarily intended to yield a categorization of “deficit” or“nondeficit” schizophrenia. Individuals meet criteria for the deficitsyndrome if they have 2 or more negative symptoms that are clinicallysignificant, and those symptoms are considered primary and stable.The SDS demonstrated good psychometric properties in the originalstudy and subsequent studies have indicated strong inter-rater reliabilityand convergent validity (Kirkpatrick et al., 1989; Fenton andMcGlashan,1992; Amador et al., 1999). Factor analytic studies suggest that the SDSitems assessing symptom severity load onto two coherent factors —

avolition and diminished emotional expressivity (Kimhy et al., 2006;Strauss et al., 2013).

The SAPS consists of 30 items that assess hallucinations, delusions,and disorganization (Andreasen, 1984). Each SAPS item is scored on a6-point scale (0–5). The Level of Function Scale (Hawk et al., 1975)mea-sures community-based functional outcome in eight areas: Duration ofHospitalization, Frequency of Social Contacts, Quality of Social Relations,

ical effort test performance in schizophrenia, Schizophr. Res. (2015),

Table 3Participant clinical characteristics by group.

3L.F. Morra et al. / Schizophrenia Research xxx (2015) xxx–xxx

Employment,WorkQuality, Symptoms, Ability toMeet ownBasicNeeds,and Fullness of Life. LOF ratings are assigned on a 0–4 scale with highscores indicating better functional outcome (and decreased symptomsfor the symptoms domain). All of the domains, except symptom severity,are rated for the last six months; symptom severity is rated for the pastmonth. Summary scores were calculated for total functional outcome,work outcome, and social outcome in the current study.

Neuropsychologicalmeasures included the (1)WideRange Achieve-ment Test — Third Edition (WRAT-3: Wilkinson, 1993), (2) WechslerAdult Intelligence Scale — Third Edition (Information, Block Design,Arithmetic, and Symbol Search subtests) (WAIS-III: Wechsler, 1997),and (3) Repeatable Battery for the Assessment of NeuropsychologicalStatus (RBANS: Randolph, 1998). The RBANS is a brief neurocognitivebattery that consists of 12 subtests measuring cognitive abilities acrossfive broad domains: immediate memory, visuospatial ability, language,attention, and delayed memory. A total score is also derived from sum-ming index scores. The RBANS has demonstrated adequate sensitivity,reliability, and validity for neurocognitive screening of individualswith schizophrenia (Gold et al., 1999; Wilk et al., 2004).

The RBANS Effort Index (EI) was calculated to evaluate insufficienteffort using standard procedures developed by Silverberg et al. (2007)that use the List Recognition and Digit Span subtests. Insufficient effortwas defined as an EI N 3.

3. Results

3.1. Insufficient effort and neuropsychological performance on the RBANS

Thirty-one patients (9.4%) failed the RBANS Effort Index (EI). Thedistribution of patients falling at each EI value was: 0 = 202 (61.2%),1 = 31 (9.4%), 2 = 36 (10.9%), 3 = 30 (9.1%), 4 = 7 (2.1%), 5 = 10(3.0%), 6 = 8 (2.4%), 7 = 5 (1.5%), and 9 = 1 (0.3%).

Given that SZ-FAIL and SZ-PASS groups significantly differed on age,age was entered as a covariate in primary analyses. As expected, sepa-rate one-way ANCOVAs indicated that the SZ-PASS and SZ-FAIL groupssignificantly differed on both Digit Span, F(1, 327) = 71.90, p b 0.001,ηp2 = 0.18, and List Recognition, F(1, 327) = 148.49, p b 0.001, ηp

2 =0.31, subtests that are used to derive the EI.

Differences in neuropsychological test performance between groupsare presented in Table 2. AMANCOVA, using age as a covariate, revealedthat patients who failed the Effort Index had lower scores on the RBANStotal score, as well as the five index scores (language, attention, visuo-spatial, immediate memory, and delayed memory) (p b .01). Effectsizes ranged from ηp

2 = 0.03 to ηp2 = 0.18. Of those indexes that do

not include subtests comprising the EI, effect sizes were the greatest

Table 2Neuropsychological performance in pass and fail groups.

SZ-PASS SZ-FAIL Test statistic, p value Effectsizes (ηp

2)

Estimated FSIQ(WAIS-III)

84.66 (16.21) 68.48 (10.35) F = 29.56, p b 0.001

WRAT-3 scaledscores

90.03 (15.16) 75.32 (17.93) F = 25.50, p b 0.001

RBANS total 72.24 (13.88) 51.13 (5.24) F = 73.56, p b 0.001 0.18Immediatememory

75.65 (17.42) 50.19 (7.78) F = 61.19, p b 0.001 0.16

Visuospatial 79.38 (17.63) 68.39 (13.48) F = 11.29, p = 0.001 0.03Language 82.88 (14.67) 69.61 (16.62) F = 31.24, p b 0.001 0.09Attention 74.47 (15.49) 51.71 (9.21) F = 72.28, p b 0.001 0.18Delayedmemory 77.40 (17.78) 49.84 (11.44) F = 70.99, p b 0.001 0.18

Notes: FSIQ=Full Scale IntelligenceQuotient,WAIS-III=Wechsler Adult Intelligence Scale,3rd Edition, WRAT-3 = Wide Range Achievement Test, 3rd Edition, RBANS = RepeatableBattery for the Assessment of Neuropsychological Status, SZ-PASS = Individuals withschizophrenia that passed the RBANSEffort Index, SZ-FAIL= Individualswith schizophreniathat failed the RBANS Effort Index. A total of 330 patients had WAIS-III data: 299 SZ-PASSand 31 SZ-FAIL, a total of 329 patients had WRAT-3 data: 298 SZ-PASS and 31 SZ-FAIL,and a total of 330 patients had RBANS data: 299 SZ-PASS and 31 SZ-FAIL.

Please cite this article as: Morra, L.F., et al., Predictors of neuropsychologhttp://dx.doi.org/10.1016/j.schres.2014.12.033

for the immediate memory (ηp2 = 0.16) and language (ηp

2 = 0.09)index scores. Consistentwith the RBANS results, the patients who failedthe Effort Index also demonstrated markedly lower IQ andWRAT read-ing scores than the patients who passed the EI.

3.2. Clinical symptoms and functional outcome

Patients in the SZ-FAIL and SZ-PASS groups differed significantly onseveral clinical symptom variables (see Table 3). The SZ-FAIL groupwasmore likely tomeet criteria for the deficit syndrome (p=0.02) and hadgreater severity of all 6 negative symptom subscales assessed by the SDS(all p-values≤ 0.05). Group differences in severity of negative symptomsubscales remained significant after controlling for SAPS total scores(all p-values ≤ 0.01), suggesting that these negative symptom effectscannot be attributed to the secondary factor of psychosis.

The SZ-FAIL group also had more severe hallucinations, bizarrebehavior, formal thought disorder, and total positive symptoms(p-values≤ 0.03) relative to the SZ-PASS group, but did not have greaterseverity of delusions (p = 0.36). Groups also differed in community-based functional outcome. More specifically, the SZ-FAIL group demon-strated significantly lower scores onmeasures of work and overall func-tioning (p=0.05 and p=0.01, respectively). In contrast, groups did notdiffer on social outcome.

To investigate whether cognitive effort was related to the volitionaland/or emotional expressivity dimensions of negative symptoms, twoindex scores were calculated using the SDS severity items. The avolitiondimension was calculated by averaging curbing of interests, diminishedsense of purpose, and diminished social drive items. The diminishedexpressivity dimension was calculated by averaging the severity scoresfor the restricted affect, diminished emotional range, and poverty ofspeech items. Analyses revealed that greater severity of both dimin-ished emotional expression and volition were significantly associatedwith higher scores on the EI (r = 0.24, p b 0.001 and r = 0.25,p b 0.001, respectively). Differences remained statistically significantfor diminished emotional expressivity (r= 0.28, p b 0.001) and volition(r = 0.30, p b 0.001) dimensions when age and SAPS total scores werepartialed out.

3.3. Predictors of insufficient effort

SDS mean severity scores, SAPS total scores, and full-scale IQ scoreswere used in a binary logistic regression analysis to predict RBANS EI

SZ-PASS SZ-FAIL Test statistic, p value

SDS global categorization 18.5% deficit 39.1% deficit χ2 = 5.51, p = 0.02SDS severity scores

Restricted affect 1.07 (1.09) 1.67 (1.37) F = 6.26, p = 0.01Diminished emotional range 0.94 (1.04) 1.54 (1.50) F = 6.77, p = 0.01Poverty of speech 0.69 (1.08) 1.29 (1.37) F = 6.41, p = 0.01Curbing of interests 1.00 (1.16) 1.92 (1.53) F = 12.83, p b 0.01Diminished sense of purpose 1.18 (1.32) 2.29 (1.55) F = 14.90, p b 0.01Diminished social drive 1.21 (1.25) 1.75 (1.442) F = 3.97, p = 0.05

Positive symptoms (SAPS)Hallucinations 1.97 (1.89) 3.05 (1.94) F = 6.52, p = 0.01Delusions 2.34 (1.67) 2.68 (1.76) F = 0.85, p = 0.36Bizarre behavior 0.75 (1.11) 1.41 (1.40) F = 6.71, p = 0.01Formal thought disorder 1.13 (1.31) 1.82 (1.87) F = 5.21, p = 0.02SAPS total score 24.05 (20.51) 34.40 (22.79) F = 4.60, p = 0.03

Functional outcomeLOF social 3.28 (2.54) 2.67 (2.14) F = 1.31, p = 0.25LOF work 2.28 (2.46) 1.25 (1.29) F = 4.05, p = 0.05LOF total score 16.20 (7.04) 12.38 (5.88) F = 6.65, p = 0.01

Notes: SZ-PASS = participants with an EI N 3, SZ-FAIL, participants with an EI ≤ 3,SDS = Schedule for the Deficit Syndrome, SAPS = Scale for the Assessment of PositiveSymptoms, LOF = Level of Functioning Scale. A total of 299 patients had SDS data: 268SZ-PASS and 31 SZ-FAIL, a total of 249 patients had SAPS data: 229 SZ-PASS and 20 SZ-FAIL, and a total of 270 patients had LOF data: 246 SZ-PASS and 24 SZ-FAIL.

ical effort test performance in schizophrenia, Schizophr. Res. (2015),

Table 5Correlations between continuous RBANS Effort Index scores and clinicalvariables.

r

Full-scale IQ −0.35⁎⁎⁎

SDS diminished expressivity 0.24⁎⁎⁎

SDS volition 0.25⁎⁎⁎

SDS total 0.27⁎⁎⁎

SAPS total 0.08LOF social −0.11LOF work −0.14⁎

LOF total −0.19⁎

Notes: SDS = Schedule for the Deficit Syndrome; SAPS = Scale for theAssessment of Positive Symptoms; LOF = Level of Functional Scale;lower scores on the LOF reflect poorer functional outcome.⁎⁎⁎ p b 0.001.⁎ p b 0.05.

4 L.F. Morra et al. / Schizophrenia Research xxx (2015) xxx–xxx

pass/fail group membership. Overall, the model correctly classified 92%of patients. The prediction model was statistically significant andaccounted for approximately 30% (Nagerkle R2) of variance in EI failure,χ2 = 34.65, p b 0.001. Negative symptoms, positive symptoms, and IQwere all significant predictors (p b 0.02) (see Table 4).

Table 5 presents correlations between dimensional RBANS EI scoresand WAIS estimated full-scale IQ, SDS symptom severity, SAPS totalscores, and LOFwork, social, and total scores. Lower effort was associat-ed with lower IQ, greater severity of volitional and expressive negativesymptoms, poorer global functional outcome, and poorer work out-come. Correlations with positive symptoms and social outcome werenonsignificant.

4. Discussion

Results indicated that only 9.4% of schizophrenia patients fell belowthe cut-off for low effort on the RBANS EI. This suggests that themajority of our sample put forth sufficient effort during neuropsycho-logical testing. The rate of patients exhibiting low effort in our study issomewhat lower than themajority of published studies, which have in-dicated that approximately 20% of peoplewith schizophrenia fail neuro-psychological effort tests. Like prior studies, we did not expect patientsfailing the EI to be malingering (i.e., feigning low performance). Rather,we expected that patients who failed the EI would display clinicallysignificantmotivational impairments that negatively impacted test per-formance. Indeed, this is what was observed— patients failing the EffortIndex had a greater severity of clinically rated negative symptoms thanpatients who passed the EI. These findings are consistent with severalprior studies reporting an association between negative symptomsand low effort (Gorissen et al., 2005; Avery et al., 2009; Strauss et al.,2014).

Most importantly,we extended the results of prior studies indicatingan association between negative symptoms and low effort by teasingapart the contributions of two critical negative symptom factors —

dimensionality and whether negative symptoms are attributed toprimary vs. secondary causes. We hypothesized that low effort wouldbe predicted by greater severity of volitional pathology and greater like-lihood ofmeeting criteria for deficit schizophrenia (i.e., primary and en-during negative symptoms). Contrary to hypotheses, low effort showeda similar magnitude of correlation with the volitional and diminishedexpressivity dimensions. This suggests that low effort is associatedwith greater severity of negative symptoms broadly defined, ratherthan increased severity of volitional pathology specifically. Results did,however, support the hypothesis that there would be a greater propor-tion of patients meeting criteria for deficit schizophrenia among thosewho failed the EI than among those who passed. This finding suggeststhat patients with primary and enduring negative symptoms are morelikely to exert low effort during cognitive testing, potentially explainingwhy many studies find that deficit schizophrenia patients perform onthe order of one standard deviation below nondeficit schizophrenia

Table 4Logistic regression model prediction RBANS Effort Index failure group.

B (SE) Wald Exp (B) 95% confidenceinterval for odds ratio

Lower Upper

IQ −0.10 (0.03)⁎⁎⁎ 11.4 0.91 0.86 0.96SAPS 0.03 (0.01)⁎ 5.6 1.03 1.01 1.05SDS 0.11 (0.04)⁎⁎ 7.9 1.11 1.03 1.20Constant 3.00 (2.05) 2.14 20.25

Notes: IQ=Wechsler Estimated Full-Scale Intelligence; SAPS= Scale for the Assessmentof Positive Symptoms Total Score; SDS = Schedule for the Deficit Syndrome averageseverity score.⁎⁎⁎ p b 0.001.⁎⁎ p b 0.01.⁎ p b 0.05.

Please cite this article as: Morra, L.F., et al., Predictors of neuropsychologhttp://dx.doi.org/10.1016/j.schres.2014.12.033

patients on standard neuropsychological tests (Cohen et al., 2007). Fur-thermore, insufficient effort was related to several other clinical vari-ables that are common secondary negative symptom causes, includingpositive and disorganized symptoms. When these variables were en-tered as covariates or partialed during correlations, negative symptomeffects persisted. Thus, although associations were observed betweenlow effort and several forms of psychopathology, results suggest thatnegative symptom effects observed are not likely to be attributable tocommon causes of secondary negative symptoms.

Interestingly, low IQ also predicted insufficient effort. Effort testresults are typically thought to be valid indicators of performance inindividuals with low or borderline intellectual functioning; however,there has been some evidence that neurological patients (e.g., epilepsy)with lower intellectual functioning fail a significantly greater number ofeffort measures than patients with higher intellectual functioning, evenin the absence of any identifiable incentive to feign cognitive impairment(Dean et al., 2008). Our results suggest that schizophrenia patients withlow IQmay also be highly susceptible to failing effort tests. Thesefindingscall into question whether effort tests are appropriate for use in schizo-phrenia patients with low IQ. It is unclear whether insufficient effortleads to low performance on IQ tests, or low IQ leads to greater rates ofeffort test failure. It is possible that the patients who failed the EI wereaffected by neurodevelopmental abnormalities years before the neuro-psychological tests were administered; results indicating that patientswho failed the EI had lower personal education and WRAT-3 readingscores than patients who passed are consistent with this notion. Thus,we suspect that either low IQ and neurodevelopmental abnormalitieslead to greater rates of effort test failure, or that the same kind of motiva-tional processes that affect performance on effort tests also present earlyin development, resulting in poor academic attainment and difficultylearning to read.

Another important finding was that insufficient effort was associat-ed with poorer neurocognitive performance across all five RBANSdomains (language, attention, visuospatial, immediate memory, anddelayed memory), as well as tests not used to compute the EI. Thesefindings are consistent with other studies indicating that low effortmay play a role in the generalized cognitive deficit that is observed inschizophrenia (Strauss et al., 2014). Furthermore, among the domainsnot used to compute the EI, there was some evidence for larger effectsizes on more cognitively demanding than less demanding tests. Thesefindings are consistent with evidence indicating that psychologicalprocesses (e.g., low expectancies for success) may play a significantrole in both neuropsychological impairment and insufficient effort(Horan et al., 2010; Couture et al., 2011; Strauss et al., 2014).

Although only a minority of our schizophrenia sample failed the EI,findings have important implications for clinical trials examining theefficacy of cognitive enhancing drugs or cognitive retraining programs.Specifically, it may be useful to screen for insufficient effort to identifypatients who should be excluded from clinical trials. The exclusion of

ical effort test performance in schizophrenia, Schizophr. Res. (2015),

5L.F. Morra et al. / Schizophrenia Research xxx (2015) xxx–xxx

such patients may increase the likelihood of observing small to moder-ate treatment effects. Alternatively, effort testing could be used toidentify patients in need of individually tailored protocols that takemotivational difficulties into account. Well-validated embedded effortmeasures, such as the RBANS EI, may offer a screening method that isboth time and cost effective.

Certain limitations and interpretative considerations should betaken into account when evaluating these findings. First, it is well-documented that various free-standing and embedded measures resultin variable rates of effort test failure (Armistead-Jehle and Hansen,2011). This may reflect differences in sensitivity and specificity acrosstests, complicating comparisons in the rate of insufficient effort acrossschizophrenia studies. Future studies could administer multiple free-standing and embedded tests to account for this problem and considerpatients to have insufficient effort only if they fail multiple tests. Fur-thermore, additional work is needed to determine the validity of usingstandardized neuropsychological tests of effort in schizophrenia. Thecurrent results pose the question of why individualswith schizophreniaare failing these tests: does poor performance on effort tests result fromhaving low IQ and poor cognition to begin with, or does low effortpartially explain poor neuropsychological test performance? Thesequestions are impossible to answer using the current study design;however, paradigms that manipulate receipt of incentives for adequateneuropsychological test performance may shed light onto whethereffort test failure reflects a cognitive deficit or motivational problem.Additionally, many tests determine insufficient effort by calculatingdiscrepancy scores among various subtests to identify statisticallyimprobable performance that is thought to signal insufficient effort.Although these procedures may validly capture insufficient perfor-mance in populations that have discrete profiles of strengths andweak-nesses (e.g., specific deficits in recall, but not recognition memory),these discrepancy based calculations may be of questionable validityfor schizophrenia patients who display a generalized neurocognitivedeficit across multiple cognitive domains and a magnitude of impair-ment that is statistically improbable in other disorders. These consider-ations are particularly important when using embedded measures suchas the RBANS EI, whichmay capture both genuine cognitive impairmentand poor motivation. Further research is clearly needed to evaluatewhether standardized neuropsychological measures of effort can bevalidly applied to individuals with schizophrenia, or whether newmea-sures are needed to evaluate insufficient cognitive effort as it occurs inthis population.

Role of funding sourceResearch supported in part by the US National Institutes of Mental Health Grant

R01MH080066 to Dr. Gold.

ContributorsGregory Strauss and James Gold designed the study. Statistical analyses andwriting of

the first draft of the manuscript were performed by Lindsay Morra, Sara Sullivan, andGregory Strauss. All authors contributed to and approved the final manuscript.

Conflict of interestAuthors have no conflicts of interest relevant to the current study.

AcknowledgmentsThe authors would like to thank the participants who completed the study, as well as

staff at the Maryland Psychiatric Research Center who contributed to data collection.

References

Amador, X.F., Kirkpatrick, B., Buchanan, R.W., Carpenter, W.T., Marcinko, L., Yale, S.A.,1999. Stability of the diagnosis of deficit syndrome in schizophrenia. Am.J. Psychiatry 156 (4), 637–639.

Andreasen, N., 1983. The Scale for the Assessment of Negative Symptoms (SANS). TheUniversity of Iowa, Iowa City, Iowa.

Andreasen, N., 1984. Scale for the Assessment of Positive Symptoms (SAPS). University ofIowa, Iowa City.

Please cite this article as: Morra, L.F., et al., Predictors of neuropsychologhttp://dx.doi.org/10.1016/j.schres.2014.12.033

Armistead-Jehle, P., Hansen, C.L., 2011. Comparison of the Repeatable Battery for theAssessment of Neuropsychological Status Effort Index and stand-alone symptomvalidity tests in a military sample. Arch. Clin. Neuropsychol. 26 (7), 592–601.

Arnold, G., Boone, K.B., Lu, P., Dean, A., Wen, J., Nitch, S., McPherson, S., 2005. Sensitivityand specificity of finger tapping test scores for the detection of suspect effort. Clin.Neuropsychol. 19 (1), 105–120.

Avery, R., Startup, M., Calabria, K., 2009. The role of effort, cognitive expectancy appraisalsand coping style in the maintenance of the negative symptoms of schizophrenia.Psychiatry Res. 167 (1–2), 36–46.

Back, C., Brauer Boone, K., Edwards, C., 1996. The performance of schizophrenics on threecognitive tests of malingering, Rey 15-Item Memory test, Rey Dot Counting, andHiscock Forced-Choice Method. Assessment 3, 449–457.

Blanchard, J.J., Cohen, A.S., 2006. The structure of negative symptoms within schizophre-nia: implications for assessment. Schizophr. Bull. 32 (2), 238–245.

Buchanan, R.W., Strauss, M.E., Kirkpatrick, B., Holstein, C., Breier, A., Carpenter Jr., W.T.,1994. Neuropsychological impairments in deficit vs nondeficit forms of schizophrenia.Arch. Gen. Psychiatry 51 (10), 804–811.

Carpenter Jr., W.T., Heinrichs, D.W., Wagman, A.M., 1988. Deficit and nondeficit forms ofschizophrenia: the concept. Am. J. Psychiatry 145 (5), 578–583.

Cohen, A.S., Saperstein, A.M., Gold, J.M., Kirkpatrick, B., Carpenter, W.T., Buchanan, R.W.,2007. Neuropsychology of the deficit syndrome: new data and meta-analysis of find-ings to date. Schizophr. Bull. 33 (5), 1201–1212.

Couture, S.M., Blanchard, J.J., Bennett, M.E., 2011. Negative expectancy appraisals anddefeatist performance beliefs and negative symptoms of schizophrenia. PsychiatryRes. 189 (1), 43–48.

Dean, A.C., Victor, T.L., Boone, K.B., Arnold, G., 2008. The relationship of IQ to effort testperformance. Clin. Neuropsychol. 22 (4), 705–722.

Dickinson, D., 2008. Digit symbol coding and general cognitive ability in schizophrenia:worth another look? Br. J. Psychiatry 193 (5), 354–356.

Dickinson, D., Harvey, P.D., 2009. Systemic hypotheses for generalized cognitive deficits inschizophrenia: a new take on an old problem. Schizophr. Bull. 35 (2), 403–414.

Dickinson, D., Iannone, V.N., Wilk, C.M., Gold, J.M., 2004. General and specific cognitivedeficits in schizophrenia. Biol. Psychiatry 55 (8), 826–833.

Dickinson, D., Ramsey, M.E., Gold, J.M., 2007. Overlooking the obvious: a meta-analyticcomparison of digit symbol coding tasks and other cognitive measures in schizophre-nia. Arch. Gen. Psychiatry 64 (5), 532–542.

Dickinson, D., Ragland, J.D., Gold, J.M., Gur, R.C., 2008. General and specific cognitivedeficits in schizophrenia: Goliath defeats David? Biol. Psychiatry 64 (9), 823–827.

Duncan, A., 2005. The impact of cognitive and psychiatric impairment of psychotic disor-ders on the test of memory malingering (TOMM). Assessment 12 (2), 123–129.

Egeland, J., Sundet, K., Rund, B.R., Asbjornsen, A., Hugdahl, K., Landro, N.I., Lund, A., Roness,A., Stordal, K.I., 2003. Sensitivity and specificity of memory dysfunction in schizophre-nia: a comparison with major depression. J. Clin. Exp. Neuropsychol. 25 (1), 79–93.

Fenton, W.S., McGlashan, T.H., 1992. Testing systems for assessment of negative symp-toms in schizophrenia. Arch. Gen. Psychiatry 49 (3), 179–184.

Fervaha, G., Foussias, G., Agid, O., Remington, G., 2014. Motivational and neurocognitivedeficits are central to the prediction of longitudinal functional outcome in schizo-phrenia. Acta Psychiatr. Scand. 130 (4), 290–299.

Fioravanti, M., Carlone, O., Vitale, B., Cinti, M.E., Clare, L., 2005. A meta-analysis of cogni-tive deficits in adults with a diagnosis of schizophrenia. Neuropsychol. Rev. 15 (2),73–95.

Gierok, S.D., Dickson, A.L., Cole, J.A., 2005. Performance of forensic and non-forensic adultpsychiatric inpatients on the test of memory malingering. Arch. Clin. Neuropsychol.20 (6), 755–760.

Gold, J.M., Queern, C., Iannone, V.N., Buchanan, R.W., 1999. Repeatable battery for theassessment of neuropsychological status as a screening test in schizophrenia I: sensi-tivity, reliability, and validity. Am. J. Psychiatry 156 (12), 1944–1950.

Gorissen, M., Sanz, J.C., Schmand, B., 2005. Effort and cognition in schizophrenia patients.Schizophr. Res. 78 (2–3), 199–208.

Hawk, A.B., Carpenter Jr., W.T., Strauss, J.S., 1975. Diagnostic criteria and five-year out-come in schizophrenia. a report from the international pilot study of schizophrenia.Arch. Gen. Psychiatry 32 (3), 343–347.

Horan, W.P., Rassovsky, Y., Kern, R.S., Lee, J., Wynn, J.K., Green, M.F., 2010. Further supportfor the role of dysfunctional attitudes in models of real-world functioning in schizo-phrenia. J. Psychiatr. Res. 44 (8), 499–505.

Horan, W.P., Kring, A.M., Gur, R.E., Reise, S.P., Blanchard, J.J., 2011. Development and psy-chometric validation of the Clinical Assessment Interview for Negative Symptoms(CAINS). Schizophr. Res. 132 (2–3), 140–145.

Hunt, S., Root, J.C., Bascetta, B.L., 2014. Effort testing in schizophrenia and schizoaffectivedisorder: validity indicator profile and test of memory malingering performancecharacteristics. Arch. Clin. Neuropsychol. 29 (2), 164–172.

Kimhy, D., Yale, S., Goetz, R.R., McFarr, L.M., Malaspina, D., 2006. The factorial structure of theschedule for the deficit syndrome in schizophrenia. Schizophr. Bull. 32 (2), 274–278.

Kirkpatrick, B., Buchanan, R.W., McKenney, P.D., Alphs, L.D., Carpenter Jr., W.T., 1989.The schedule for the deficit syndrome: an instrument for research in schizophrenia.Psychiatry Res. 30 (2), 119–123.

Kirkpatrick, B., Buchanan, R.W., Ross, D.E., Carpenter, W.T., 2001. A separate diseasewithin the syndrome of schizophrenia. Arch. Gen. Psychiatry 58 (2), 165–171.

Kirkpatrick, B., Strauss, G.P., Nguyen, L., Fischer, B.A., Daniel, D.G., Cienfuegos, A., Marder,S.R., 2011. The brief negative symptom scale: psychometric properties. Schizophr.Bull. 37 (2), 300–305.

Kraepelin, E., 1919. Dementia praecox and paraphrenia. B. RM, Trans., New York.Moore, R.C., Davine, T., Harmell, A.L., Cardenas, V., Palmer, B.W., Mausbach, B.T., 2013.

Using the Repeatable Battery for the Assessment of Neuropsychological Status(RBANS) effort index to predict treatment group attendance in patients with schizo-phrenia. J. Int. Neuropsychol. Soc. 19 (2), 198–205.

ical effort test performance in schizophrenia, Schizophr. Res. (2015),

6 L.F. Morra et al. / Schizophrenia Research xxx (2015) xxx–xxx

Pivovarova, E., Rosenfeld, B., Dole, T., Green, D., Zapf, P., 2009. Are measures of cognitiveeffort and motivation useful in differentiating feigned from genuine psychiatricsymptoms? Int. J. Forensic Mental Health 8, 271–278.

Randolph, C., 1998. The repeatable battery for the assessment of neuropsychological sta-tus manual. The Psychological Corporation, San Antonio, TX.

Reichenberg, A., Harvey, P.D., 2007. Neuropsychological impairments in schizophrenia:integration of performance-based and brain imaging findings. Psychol. Bull. 133(5), 833–858.

Schroeder, R.W., Marshall, P.S., 2011. Evaluation of the appropriateness of multiplesymptom validity indices in psychotic and non-psychotic psychiatric populations.Clin. Neuropsychol. 25 (3), 437–453.

Silverberg, N.D., Wertheimer, J.C., Fichtenberg, N.L., 2007. An effort index for the Repeat-able Battery For The Assessment Of Neuropsychological Status (RBANS). Clin.Neuropsychol. 21 (5), 841–854.

Strauss, G.P., Hong, L.E., Gold, J.M., Buchanan, R.W., McMahon, R.P., Keller, W.R., Fischer,B.A., Catalano, L.T., Culbreth, A.J., Carpenter, W.T., Kirkpatrick, B., 2012. Factor struc-ture of the Brief Negative Symptom Scale. Schizophr. Res. 142 (1–3), 96–98.

Please cite this article as: Morra, L.F., et al., Predictors of neuropsychologhttp://dx.doi.org/10.1016/j.schres.2014.12.033

Strauss, G.P., Horan, W.P., Kirkpatrick, B., Fischer, B.A., Keller, W.R., Miski, P., Buchanan,R.W., Green, M.F., Carpenter Jr., W.T., 2013. Deconstructing negative symptoms ofschizophrenia: avolition–apathy and diminished expression clusters predict clinicalpresentation and functional outcome. J. Psychiatr. Res. 47 (6), 783–790.

Strauss, G.P., Morra, L.F., Sullivan, S.K., Gold, J.M., 2014. The role of low cognitive ef-fort and negative symptoms in neuropsychological impairment in schizophrenia.Neuropsychology (Advance online publication).

Wechsler, D., 1997. Wechsler adult intelligence scale. 3rd edition. The Psychological Cor-poration, San Antonio, TX.

Wilk, C.M., Gold, J.M., Humber, K., Dickerson, F., Fenton, W.S., Buchanan, R.W., 2004. Briefcognitive assessment in schizophrenia: normative data for the Repeatable Battery forthe Assessment of Neuropsychological Status. Schizophr. Res. 70 (2–3), 175–186.

Wilkinson, G.S., 1993. The Wide Range Achievement Test: Manual. third ed. Wide RangeInc., Wilmington, DE.

ical effort test performance in schizophrenia, Schizophr. Res. (2015),